423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2026-07-07 09:35:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

[v1] support training with fsdp2 (#9773)

Browse Source 
Co-authored-by: frozenleaves <frozen@Mac.local>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
This commit is contained in:
浮梦
2026-01-25 19:41:58 +08:00
committed by GitHub
parent 641bfdd482
commit f9f11dcb97
15 changed files with 801 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llamafactory/v1/accelerator/interface.py
View File

@@ -145,7 +145,7 @@ class DistributedInterface:
timeout = config.get("timeout", 18000)
if self._is_distributed:
init_process_group(timeout=timedelta(seconds=timeout))
init_process_group(timeout=timedelta(seconds=timeout), backend=helper.get_process_group_backend())
self.model_device_mesh = init_device_mesh(
device_type=self.current_device.type,
mesh_shape=self.strategy.model_mesh_shape,